章城皓,吴赋章,杨军,别芳玫,张国航,黄喆,林方舟
ZHANG Chenghao,WU Fuzhang,YANG Jun,BIE Fangmei,ZHANG Guohang,HUANG Zhe,LIN Fangzhou

• 1:武汉大学电气与自动化学院
• 2:国网湖北省电力有限公司经济技术研究院

摘要(Abstract)：

EV（电动汽车）充电负荷预测的准确性对电力系统的稳定运行和资源的合理配置起着重要作用。然而，充电功率序列呈现出显著的日内与日间周期性波动、随机性波动以及非线性特征，传统预测方法难以有效协调周期性与随机性之间的复杂关系，导致预测精度受限。针对上述问题，提出一种LaST-BO（潜在周期-缓变性表示学习框架和贝叶斯优化）模型。该模型借助LaST模型的变分推理机制，解耦充电功率序列的周期性与缓变性成分；同时运用贝叶斯优化对模型超参数进行智能寻优，增强模型适应性。基于实际充电站数据进行了仿真实验，并与多种传统模型及深度学习模型进行对比分析，结果表明LaST-BO模型有着较高的预测精度，并且能够在线应用。
Accurate forecasting of electric vehicle(EV) charging load is essential for the stable operation of power systems and the efficient allocation of resources. However, charging power sequences exhibit pronounced intradaily and interdaily periodic fluctuations, random volatility, and nonlinear characteristics. Conventional forecasting methods struggle to adequately capture the complex interplay between periodicity and randomness, leading to limited forecasting accuracy. In response to these challenges, this paper proposes a hybrid model that integrates learning latent cyclical-gradual change representations for time series forecasting with Bayesian optimization(LaST-BO). This model leverages the variational inference mechanism of the LaST framework to disentangle the periodic and gradual components of charging power sequence. Moreover, Bayesian optimization is employed to intelligently tune the model hyperparameters, thereby enhancing its adaptability. Simulation experiments using real-world charging station data were conducted, and comparisons with multiple conventional and deep-learning models demonstrate that the LaST-BO hybrid model achieves higher forecasting accuracy and is suitable for online deployment.

关键词(KeyWords)： 电动汽车;功率预测;解耦表示学习;贝叶斯优化;变分推理;趋势分解
electric vehicle;power forecasting;disentangled representation learning;Bayesian optimization;variational inference;trend decomposition

Abstract:

Keywords:

基金项目(Foundation): 国家电网有限公司科技项目(521500250018-054-ZN)

作者(Author): 章城皓,吴赋章,杨军,别芳玫,张国航,黄喆,林方舟
ZHANG Chenghao,WU Fuzhang,YANG Jun,BIE Fangmei,ZHANG Guohang,HUANG Zhe,LIN Fangzhou

DOI: 10.19585/j.zjdl.202605005

参考文献(References)：

• [1]刘洪，张旭，刘畅，等.考虑充电设施充裕性的电动私家车出行与充电需求时序交互分析[J].中国电机工程学报，2018,38(18):5469-5478.LIU Hong,ZHANG Xu,LIU Chang,et al.Timing interactive analysis of electric private vehicle traveling and charging demand considering the sufficiency of charging facilities[J].Proceedings of the CSEE,2018,38(18):5469-5478.
• [2]陈丽丹，聂涌泉，钟庆.基于出行链的电动汽车充电负荷预测模型[J].电工技术学报，2015,30(4):216-225.CHEN Lidan,NIE Yongquan,ZHONG Qing. A model for electric vehicle charging load forecasting based on trip chains[J].Transactions of China Electrotechnical Society,2015,30(4):216-225.
• [3]邵尹池，穆云飞，余晓丹，等.“车-路-网”模式下电动汽车充电负荷时空预测及其对配电网潮流的影响[J].中国电机工程学报，2017,37(18):5207-5219.SHAO Yinchi,MU Yunfei,YU Xiaodan,et al.A spatialtemporal charging load forecast and impact analysis method for distribution network using EVs-trafficdistribution model[J].Proceedings of the CSEE,2017,37(18):5207-5219.
• [4]茆美琴，葛星，杨铖，等.计及用户偏好的电动汽车集群参与调频市场优化控制[J].电力系统自动化，2025,49(14):130-140.MAO Meiqin,GE Xing,YANG Cheng,et al.Optimal control of electric vehicle cluster participating in frequency regulation market considering user preferences[J]. Automation of Electric Power Systems,2025,49(14):130-140.
• [5]谭楠玲.电动汽车有序充电管理优化策略研究[D].重庆：重庆三峡学院，2023.TAN Nanling. Research on optimal strategy for orderly charging management of electric vehicles[D].Chongqing:Chongqing Three Gorges University,2023.
• [6]郑颖颖，缪新义，王晖，等.支持时变需求响应激励价格的电动汽车优化充电策略[J].电力系统自动化，2025,49(9):96-106.ZHENG Yingying,MIAO Xinyi,WANG Hui,et al.Optimal charging strategies for electric vehicles supporting time-varying demand response incentive Price[J].Automation of Electric Power Systems,2025,49(9):96-106.
• [7]胡枭，张泽朕，杨家全，等.融合路网-气象-日期多特征信息的电动汽车充电负荷预测[J].电力建设，2025,46(9):57-70.HU Xiao,ZHANG Zezhen,YANG Jiaquan,et al.Electric vehicle charging load forecasting by integrating multifeature information of road network,meteorology,and date[J].Electric Power Construction,2025,46(9):57-70.
• [8]沈筱琦，方鑫，谭林林，等.基于居民出行模拟的电动汽车负荷时空分布预测[J].电力工程技术，2024,43(3):130-139.SHEN Xiaoqi,FANG Xin,TAN Linlin,et al.Prediction of spatio-temporal distribution of electric vehicle load based on residential travel simulation[J].Jiangsu Electrical Engineering,2024,43(3):130-139.
• [9]张琳娟，李文峰，许长清，等.计及充电排队与实时SOC的电动汽车充电负荷时空分布预测[J].电力建设，2025,46(8):54-66.ZHANG Linjuan,LI Wenfeng,XU Changqing,et al.Spatial-temporal distribution prediction of electric vehicle charging load considering charging behavior and real-time SOC[J].Electric Power Construction,2025,46(8):54-66.
• [10]郭鑫喆，王业琴，王超，等.基于多目标模态分解与NAHL神经网络的电动汽车充电负荷预测方法[J].电测与仪表，2025,62(3):20-29.GUO Xinzhe,WANG Yeqin,WANG Chao,et al.Electric vehicle charging load prediction method based on multiobjective modal decomposition and NAHL neural network[J].Electrical Measurement&Instrumentation,2025,62(3):20-29.
• [11]陈晓华，吴杰康，张勋祥，等.基于鹈鹕优化算法优化广义回归神经网络的电动汽车充电负荷短期预测[J].山东电力技术，2024,51(7):1-9.CHEN Xiaohua,WU Jiekang,ZHANG Xunxiang,et al.Electric vehicle charging load short-term prediction based on generalized regression neural network optimized by pelican optimization algorithm[J].Shandong Electric Power,2024,51(7):1-9.
• [12]丁乐言，柯松，张帆，等.考虑出行需求和引导策略的电动汽车充电负荷预测[J].电力建设，2024,45(6):10-26.DING Leyan,KE Song,ZHANG Fan,et al.Forecasting of electric-vehicle charging load considering travel demand and guidance strategy[J]. Electric Power Construction,2024,45(6):10-26.
• [13]陈楷，高丽萍，闫珺，等.基于分数阶控制的充电站优化控制方法[J].电力电容器与无功补偿，2025,46(5):54-64.CHEN Kai,GAO Liping,YAN Jun,et al.Optimal control method for charging stations based on fractional-order control[J]. Power Capacitor&Reactive Power Compensation,2025,46(5):54-64.
• [14]龙雪梅，杨军，吴赋章，等.考虑路网-电网交互和用户心理的电动汽车充电负荷预测[J].电力系统自动化，2020,44(14):86-93.LONG Xuemei,YANG Jun,WU Fuzhang,et al.Prediction of electric vehicle charging load considering interaction between road network and power grid and user’s psychology[J].Automation of Electric Power Systems,2020,44(14):86-93.
• [15]李宏伟，宋玉峰，李帅兵，等.基于ArcGIS路网结构与交通拥挤度分析的电动汽车充电负荷预测方法[J].电网技术，2025,49(5):1920-1930.LI Hongwei,SONG Yufeng,LI Shuaibing,et al.Electric vehicle charging load prediction based on ArcGIS road network structure and traffic congestion analysis[J]. Power System Technology,2025,49(5):1920-1930.
• [16]毛建斌，杨少兵，杨湘彦，等.基于有限混合模型的规模化电动汽车充电负荷在线预测方法[J].电网技术，2025,49(5):1931-1940.MAO Jianbin,YANG Shaobing,YANG Xiangyan,et al.Charging load online forecasting method based on finite mixture model for large-scale electric vehicles[J]. Power System Technology,2025,49(5):1931-1940.
• [17]郭星辰.规模化电动汽车充电负荷预测及需求响应调控技术研究[D].西安：西安理工大学，2024.GUO Xingchen.Research on charging load forecasting and demand response regulation technology of large-scale electric vehicles[D].Xi’an:Xi’an University of Technology,2024.
• [18]张欣，李林.基于随机森林的季节性电动汽车充电负荷预测[J].软件工程，2024,27(11):11-14.ZHANG Xin,LI Lin. Seasonal electric vehicle charging load prediction based on random forest[J].Software Engineering,2024,27(11):11-14.
• [19]赵伟博，范旭东，孙胜博，等.电动汽车充电负荷场景化分析与超短期预测方法[J].电力信息与通信技术，2025,23(2):28-37.ZHAO Weibo,FAN Xudong,SUN Shengbo,et al. A scenario-based analysis and ultra-short-term forecasting method for electric vehicle charging load[J]. Electric Power Information and Communication Technology,2025,23(2):28-37.
• [20]苏粟，李玉璟，夏明超，等.基于时空耦合特性的充电站运行状态预测[J].电力系统自动化，2022,46(3):23-32.SU Su,LI Yujing,XIA Mingchao,et al. Operation state prediction of charging station based on spatio-temporal coupling characteristics[J].Automation of Electric Power Systems,2022,46(3):23-32.
• [21]黄健，陈建红，何剑杰，等.基于GCN-LSTM的电动汽车负荷预测方法[J].浙江电力，2024,43(12):59-67.HUANG Jian,CHEN Jianhong,HE Jianjie,et al.An EV load forecasting method for using GCN-LSTM[J].Zhejiang Electric Power,2024,43(12):59-67.
• [22]胡博，张鹏飞，黄恩泽，等.基于图WaveNet的电动汽车充电负荷预测[J].电力系统自动化，2022,46(16):207-213.HU Bo,ZHANG Pengfei,HUANG Enze,et al. Graph WaveNet based charging load forecasting of electric vehicle[J].Automation of Electric Power Systems,2022,46(16):207-213.
• [23]闫威，李南，沈月秀，等.基于CNN-GAN与半监督回归的电动汽车充电负荷预测[J].浙江电力，2023,42(2):83-89.YAN Wei,LI Nan,SHEN Yuexiu,et al.Electric vehicle charging load forecasting based on CNN-GAN and semisupervised regression[J].Zhejiang Electric Power,2023,42(2):83-89.
• [24]郝建华.基于深度学习的时间序列预测算法研究[D].济南：山东师范大学，2024.HAO Jianhua. Research on time series forecasting algorithm based on deep learning[D].Jinan:Shandong Normal University,2024.
• [25]刘建建，周军，余卫东，等.基于超参数优化LSTM的声波测井曲线生成技术[J].石油物探，2024,63(5):1061-1074.LIU Jianjian,ZHOU Jun,YU Weidong,et al.Acoustic log curve reconstruction based on hyperparameter optimized LSTM[J].Geophysical Prospecting for Petroleum,2024,63(5):1061-1074.
• [26]WANG Z Y,XU X,ZHANG W F,et al.Learning latent seasonal-trend representations for time series forecasting[C]//NEURIPS:2022:38775-38787.